catoc

10 ms·

121.
by catoc 10mo ago
If you ask two humans to explain a problem to you, and human 1 takes an hour to explain what human 2 explained in 5 minutes… everyone would consider human 1 LESS ‘productive’ than human 2. But what if human 2 was wrong? What if both were wr
122.
by catoc 10mo ago
It works, weber you like it or not (⌐⊙_⊙) (Slightly more seriously, the diamagnetic properties of Sn would in actuality very much interfere with the B1 field modulation of the (f)MRI sequence; and disturb the local B0 homogeneity; and thus
123.
by catoc 10mo ago
If you don’t want the EEG to capture your brainwaves you can wear a tin foil hat to lead the magnetic field astray.
124.
by catoc 10mo ago
Always afraid that auto dismiss results in auto allow-all on thus type of dialog.
125.
by catoc 10mo ago
I would have definitely left normally. Just wanted to see the site. And I know this also likely not Neil's idea of fun, and mostly the silly EU rules that are to blame but still, dialogs without a directly available "refuse all&qu
126.
by catoc 10mo ago
No.fun in the cookie dialogue. Had to click 26 (sic!) switches to opt out of being tracked.
127.
by catoc 11mo ago
“we tell models what they can and can't say.” Thus introducing our worldly our biases
128.
by catoc 11mo ago
Bias is a reflection of real world values. The problem is not with the AI model but with the world we created. Fix the world, ‘fix’ the model.
129.
by catoc 11mo ago
That makes sense. Maybe just simply state it? E.g, instead of “you solved paprika”: “You got 5 slabs right; 10 more to prove you really solved the rule” (Being better versed in making games than I am, you can likely come up with more enti
130.
by catoc 11mo ago
Very cool! I solved the first puzzle: -Congratulations! -You solved Paprika with 18 slabs But this was unclear: -You've solved 0 puzzles! -Reveal Rule -Next Puzzle -View Archive -You still have 2 guesses left. Finish guessing before re
131.
by catoc 11mo ago
Very cool - thanks! On mobile I’d prefer to be able to just keep scrolling instead of having to tap ‘next page’, scroll to the top, then scroll down again
132.
by catoc 11mo ago
Sure even though, as most others, my server supports TLS, having your email not leave gmail at all may be slightly more secure. Part of the point however was that when either server or receiver is using Gmail, your possibly confidential ema
133.
by catoc 11mo ago
Exactly, I’d never use Gmail for anything sensitive. Even for just personal emails I use my own mailserver. (And again, for truly sensitive stuff I don’t use email at all)
134.
by catoc 11mo ago
I believe you - convenience gets picked over security all the time
135.
by catoc 11mo ago
Yes I know this was about wrong delivery address (person with same name, wrong account); the point is that email is not completely secure - certainly not for very sensitive (legal) content
136.
by catoc 11mo ago
Yeah, that exactly my point - no idea why I’m being downvoted on this
137.
by catoc 11mo ago
Even if the law firm uses a Gmail account - which most of course don’t - Google still has access to your sensitive legal email content. (And that’s apart from the meta data leaking)
138.
by catoc 11mo ago
Hmm, yes, I don’t know what I want, but I’m sure we’ll hash it out together
139.
by catoc 11mo ago
Law firms that send very sensitive legal documents over email… #sigh I’d switch firms immediately if that’s their level of opsec awareness
140.
by catoc 11mo ago
> “You and I have very different ideas of “random” I think.” Indeed, not ‘random’. With respect to iOS26 what word should one use? Premeditated? Deliberate? Maliciously?
141.
by catoc 1y ago
Reinventing the wheel for Formula 1 driving…
142.
by catoc 1y ago
Was the HN-post title also hashed? (It’s inconstitent with the actual title)
143.
by catoc 1y ago
No, this was not an off-hand remark. He made a whole story comparing the profession to the coyote from road runner “they’ve already run of the cliff but don’t even realize it”. It was callous, and showed a total ignorance of the fact that m
144.
by catoc 1y ago
As a doctor and full stack engineer you’d have a perfect future ahead of you in radiology - the profession will not go away, but will need doctors who can bridge the full medical-tech range
145.
by catoc 1y ago
You’re right - TIL
146.
by catoc 1y ago
I was thinking the exact same thing: if you don’t change the weights, use identical “temperature” etc, the same prompt will yield the same output. Under the hood it’s still deterministic code running on a deterministic machine
147.
by catoc 1y ago
Breaking encryption to stop criminals and CSAM-sharing bastards does not work. Breaking encryption will only harm honest, law-abiding citizens. Criminals will just use “illegal” real encryption. It’s easy, the implementation details are eve
148.
by catoc 1y ago
From what I understand it’s just a WKWebView. I’m trying to understand why the embedding of a WKWebView poses additional risk because it’s embedded in iTerm? (aside from the suggested general earlier security mishaps).
149.
by catoc 1y ago
What is the additional security hazard you see?
150.
by catoc 1y ago
In my experience Dropbox is actually really good at syncing data between devices. It’s fast. It’s way more reliable than iCloud, and for “simply” keeping folders in sync just “simply” the best - for simple user requirements simplicity and

More ›